1. Field of the Invention
The present invention relates to the monitoring of oxygen concentration and, more particularly, to novel, improved methods and apparatus for monitoring the concentration of oxygen in respiratory and other gases.
2. State of the Art
The most common cause of anesthetic and ventilator related mortality and morbidity is inadequate delivery of oxygen to a patient's tissues. Therefore, the monitoring of static inspired oxygen concentration has long been a safety standard of practice to ensure detection of hypoxic gas delivery to patients undergoing surgery and to those on mechanical ventilators and receiving supplemental oxygen therapy. However, monitoring the static inspired fraction of inhaled oxygen does not always guarantee adequate oxygen delivery to the tissues because it is the alveolar oxygen concentration that eventually enriches the blood delivered to the cells.
It is this alveolar gas phase that is interfaced with pulmonary perfusion which, in turn, is principally responsible for controlling arterial blood gas levels. It is very important that the clinician know the blood gas levels (partial pressure) of oxygen (pO2) and carbon dioxide (pCO2) as well as the blood pH. Blood gas levels are used as an indication of incipient respiratory failure and in optimizing the settings on ventilators. In addition, blood gas levels can detect life-threatening changes in an anesthetized patient undergoing surgery:
The traditional method for obtaining arterial blood gas values is highly invasive. A sample of arterial blood is carefully extracted and the partial pressure of the gases is measured using a blood gas analyzer. Unfortunately, arterial puncture has significant limitations:    (1) arterial puncture requires a skilled health care provider and it carries a significant degree of patient discomfort and risk, (2) handling the blood is a potential health hazard to the health care provider, (3) significant delays are often encountered before results are obtained, and (4) measurements can only be made intermittently.
Noninvasive methods for estimating blood gas levels are available. Such methods include the use of capnography (CO2 analysis). These methods employ fast gas analyzers at the patient's airway and give a graphic portrayal of breath-by-breath gas concentrations and, therefore, can measure the peak exhaled (end tidal) concentrations of the respective respired gases. Although gradients can occur between the actual arterial blood gas levels and the end tidal values, this type of monitoring is often used as a first order approximation of the arterial blood gas values.
Other techniques have been utilized for assessing patient blood gas levels with mixed results. Transcutaneous sensors measure tissue pO2 and pCO2 diffused through the heated skin surface. This type of sensor has a number of practical limitations including a slow speed of response and difficulty of use.
Pulse oximetry is widely used to measure the percentage of hemoglobin that is saturated with oxygen. Unfortunately, it does not measure the amount of dissolved oxygen present nor the amount of oxygen carried by the blood when the hemoglobin levels are reduced. This is important because low hemoglobin levels are found when there is a significant blood loss or when there is insufficient red blood cell information. In addition, pulse oximeter readings are specific to the point of contact, which is typically the finger or ear lobe, and may not reflect the oxygen level of vital organs during conditions such as shock or hypothermia.
Oxygraphy measures the approximate concentration of oxygen in the vital organs on a breath-by-breath basis and can quickly detect imminent hypoxemia due to decreasing alveolar oxygen concentration. For example, during hypoventilation, end tidal oxygen concentration changes more rapidly than does end tidal carbon dioxide. During the same conditions, pulse oximetry takes considerably longer to respond. Fast oxygen analysis (oxygraphy) can also readily detect inadvertent administration of hypoxic gas mixtures.
Oxygraphy reflects the balance of alveolar O2 available during inspiration minus the O2 uptake secondary to pulmonary perfusion. An increasing difference between inspiratory and end tidal oxygen values is a rapid indicator of a supply/demand imbalance which could be a result of changes in ventilation, diffusion, perfusion and/or metabolism of the patient. This imbalance must be quickly corrected because failure to meet oxygen demand is the most common cause of organ failure, cardiac arrest, and brain damage. Oxygraphy provides the earliest warning of the development of an impending hypoxic episode.
Oxygraphy has also been shown to be effective in diagnosing hypovolemic or septic shock, air embolism, hyperthermia, excessive positive-end expiratory pressure (PEEP), cardiopulmonary resuscitation (CPR) efficacy, and even cardiac arrest. During anesthesia, oxygraphy is useful in providing a routine monitor of preoxygenation (denitrogenation). It especially contributes to patient safety by detecting human errors, equipment failures, disconnections, misconnections, anesthesia overdoses, and esophageal intubations.
Combining the breath-by-breath analysis of oxygen with the measurement of airway flow/volume as outlined in U.S. Pat. Nos. 5,347,843 and 5,379,650 gives another dimension to the clinical utility of oxygraphy. This combination parameter, known as oxygen consumption (VO2), provides an excellent overall patient status indicator. Adequate cardiac output, oxygen delivery, and metabolic activity are all confirmed by oxygen consumption because all of these physiological processes are required for oxygen consumption to take place. Oxygen consumption is also useful in predicting ventilator weaning success.
A metabolic measurement (calorimetry) includes determination of a patient's energy requirements (in calories per day) and respiratory quotient (RQ). Interest in the measurement of caloric requirements has closely paralleled the development of nutritional support. For example, the ability to intravenously provide all the necessary nutrition to critically ill patients has only been accomplished within the last 25 years. Along with the realization that we need to feed patients has come the need to know how much to feed them, what kind of nutrients (carbohydrates, lipids, protein) to feed them, and in what ratio the nutrients need to be supplied. The only true way to measure the caloric requirements of patients and to provide a noninvasive quality assessment of their response to nutrition is with indirect calorimetry. Airway O2 consumption and CO2 production can be measured noninvasively and provide a basis for the computations needed for a measurement of indirect calorimetry, a direct measurement of the metabolic status of the patient, and the patients' respiratory quotients.
With the above clinical need in mind, it is important to ensure that clinicians have the proper equipment to monitor breath-by-breath oxygen. While there are adequate devices for measuring static levels of oxygen, the measurement of breath-by-breath (fast) airway oxygen concentration requires more sophisticated instruments. Very few of these devices can be directly attached to the patient airway. Instead, most require the use of sampling lines to acquire the gas and send it to a remote site for analysis. Fast airway oxygen monitors are typically large, heavy, fragile instruments that consume considerable power. They must sample airway gases via a small bore plastic tube (sidestream) and remotely detect the oxygen gas as it passes from the airway to the sensor. The problems associated with this type of gas sampling are well known. Gas physics dictates painstaking, careful measurements because water vapor concentration pressure and temperature can vary within the patient's airway and gas sample line. The presence of water and mucous create problems for long term patency of the sample tube. Also, the sample line acts like a low pass filter and affects the fidelity of the measurement. Finally, the pressure variable delay introduced by the sample line creates difficulty in accurately synchronizing the airway flow and oxygen concentration signals required to calculate oxygen consumption.
On-airway (mainstream) monitoring of oxygen has the potential to solve all of the above problems, especially when breath-by-breath oxygen consumption measurements are made. However, most of the available fast oxygen sensors are simply too big, too heavy, too fragile, and/or otherwise not suited to be placed in line with a patient's breathing tube.
There are various other technologies which have been employed in monitoring airway oxygen concentration. Some of the most widely used are electrochemical sensors. These fall into two basic categories: polarographic cells and galvanic cells. These cells produce an electric current proportional to the number of oxygen molecules which diffuse across a membrane. The advantages of these types of sensors are simplicity and low cost. The disadvantages of these types of sensors include limited lifetime (chemistry depletes) and slow response (not breath-by-breath). In some cases, these cells have demonstrated sensitivity to certain anesthetic agents, which introduces inaccuracies into the oxygen concentration measurement. Generally, this type of sensor is too large to attach to the patient airway.
There have been a few reported developments where electrochemical cell membranes were improved to enable faster response. There are also silicon micromachined cells using the principle of “Back Cell” electrochemical technology. Their time response approaches 150 ms but they appear to be subject to the typical problems of this type of cell (i.e., stability and calibration).
Another popular medical oxygen sensor is the paramagnetic type. This sensor uses the strong magnetic property of oxygen as a sensing mechanism. There are two basic types of paramagnetic cells: static and dynamic. The static type is a dumbbell assembly suspended between the poles of a permanent magnet. The magnetic forces of the surrounding oxygen molecules cause a torsional rotation of the dumbbell which can be sensed optically and employed as a measure of oxygen concentration. The dynamic type (see U.S. Pat. No. 4,633,705) uses a magneto-acoustic approach. This requires a gas sample and a reference gas that are mixed within an electromagnetic field. When the field is switched on and off, a pressure signal proportional to the oxygen content is generated. The signal can be detected by a differential microphone. The advantages of the paramagnetic sensor are good linearity and stability. The dynamic type has an inherently faster response than the static type. Both types are subject to mechanical vibration, and the dynamic type has the disadvantage of requiring a reference gas. Neither type is suitable for on-airway applications.
Zirconium oxide cells are frequently used in the automotive industry to measure oxygen concentration. The cell is constructed from a solid electrolyte tube covered by platinum electrodes. When heated to approximately 800 degrees C., a voltage proportional to the logarithm of the ratio between a sample gas and a reference gas is generated. The advantages of this sensor are wide dynamic range, very fast response, and simplicity. The high cell temperature is clearly a disadvantage as is power consumption. Also, the cell is degraded in the presence of anesthetic agents. Clearly, this type of cell cannot be used on a patient airway.
Ultraviolet absorption uses the principle that oxygen exhibits absorption properties in the ultraviolet part of the electromagnetic spectrum (about 147 nm). This technique has been used in several medical applications but has never been reduced to commercial viability. There are numerous technical difficulties which make this a difficult technique for on-airway applications.
Mass spectrometers spread ionized gas molecules into a detectable spectrum according to their mass-to-charge ratios and can accordingly be used to measure oxygen concentration. These instruments are generally large assemblies with ionizing magnets and high vacuum pumps. The advantages of mass spectrometers include high accuracy, multi-gas analysis capability, and rapid response. The disadvantages include high cost, high power consumption, and large size. Mass spectrometers are not suitable for on-airway applications.
Raman scattering spectrometers (as described in U.S. Pat. No. 4,784,486) can also be used to measure oxygen concentration. These devices respond to photons emitted by the collision of a photon with an oxygen molecule. A photon from a high-power laser loses energy to the oxygen molecule and is re-emitted at a lower energy and frequency. The number of photons re-emitted at the oxygen scattering wavelength is proportional to the number of oxygen molecules present. Like mass spectrometers, Raman spectrometers have multi-gas analysis capability and rapid response time. Disadvantages include large size and power consumption. Therefore, Raman scattering photometers are not suitable for on-airway applications.
Visible light absorption spectrometers (as described in U.S. Pat. Nos. 5,625,189 and 5,570,697) utilize semiconductor lasers that emit light at near 760 nm, an area of the spectrum comprised of weak absorption lines for oxygen. With sophisticated circuitry, the laser can be thermally and/or electronically tuned to the appropriate absorption bands. The amount of energy absorbed is proportional to the number of oxygen molecules present. The advantages of this system are precision, fast response, and no consumable or moving parts. The disadvantages include somewhat fragile optical components, sensitivity to ambient temperature shifts, and a long gas sample path length. While there have been attempts to utilize this technology in an on-airway configuration, no commercially viable instruments have so far been available.
Luminescence-quenching has also been proposed as a technique for measuring oxygen concentration. In this approach, a sensor contacted by the gases being monitored is excited into luminescence. This luminescence is quenched by the oxygen in the monitored gases. The rate of quenching is related to the partial pressure of oxygen in the monitored gases, and that parameter can accordingly be used to provide an indication of the oxygen in the monitored gases. However, the prior art does not disclose an oxygen concentration monitor employing luminescence-quenching which addresses the problems associated with this type of measurement device in any practical application. These problems include: photo-degradation-associated and other instabilities of the sensor, low signal level, noise leading to difficulties in assessing the decay of sensor luminescence, acceptably fast response times, thermal drift of the sensor, reproducibility of the sensors, inaccuracies attributable to stray light reaching the data photodetector, and the need for light weight, ruggedness, and low power consumption. Disclosed in copending applications Ser. Nos. 09/128,918 and 09/128,897, both filed Aug. 4, 1998, are devices for monitoring oxygen concentration in gaseous mixtures which differ from the majority of the oxygen monitors described above in that they are compact, lightweight, and otherwise suited for on-airway mainstream monitoring of the oxygen concentration in a person's respiratory gases. These monitoring devices utilize the fast (or breath-by-breath) approach to oxygen concentration monitoring with the quenching of a luminescent dye being used in determining the concentration of oxygen in the gases being monitored.
Fast (breath-by-breath) monitoring of end tidal oxygen is an important diagnostic tool because, as examples only:                1. It is a sensitive indicator of hypoventilation.        2. It aids in rapid diagnosis of anesthetic/ventilation mishaps such as (a) inappropriate gas concentration, (b) apnea, and (c) breathing apparatus disconnects.        3. End tidal oxygen analysis reflects arterial oxygen concentration.        4. Inspired-expired oxygen concentration differences reflect adequacy of alveolar ventilation. This is useful for patients undergoing ECMO (Extracorporeal Membrane Oxygenation) or nitric oxide therapies.        5. When combined with a volume flow device (e.g. a pneumotach), VO2 (oxygen consumption) can be determined. Oxygen consumption is a very useful parameter in determining (a) oxygen uptake during ventilation or exercise, (b) respiratory exchange ratio or RQ (respiratory quotient) and (c) general patient metabolic status.        
The novel sensor devices disclosed in the copending applications locate a luminescent chemical in the patient airway. Modulated visible light excites the chemical and causes it to luminesce. The lifetime of the luminescence is proportional to the amount of oxygen present. A transducer containing a photodetector and associated electronic circuitry measures decay time and relates the measured parameter to the ambient oxygen partial pressure.
The transducer device is small (<1 cubic inch), lightweight (less than 1 ounce), and does not contain moving parts. It utilizes visible light optoelectronics and consumes minimal power (system power less than 2 watts). The unit warms up in less than 30 seconds, which is advantageous in on-airway applications because of the need to take prompt remedial action if a change occurs in a patient's condition reflected in a change in respiratory oxygen concentration. The assembly does not require any significant optical alignment and is very rugged (capable of being dropped from 6 feet without affecting optical alignment or otherwise damaging the device).
The principles of the inventions disclosed in the copending applications can be employed to advantage in sidestream (sampling) type systems as well as in mainstream systems. This is important because some gas analysis systems, such as anesthetic analyzers, employ sidestream techniques to acquire their gas sample.
A typical transducer unit is easy to calibrate, is stable (±2 torr over 8 hours at a 21 percent oxygen concentration), and has a high resolution (0.1 torr) and a wide measurement range (oxygen concentrations of 0 to 100 percent). Response to changing oxygen concentrations is fast (<100 ms for oxygen concentrations of 10-90 percent at flow rates ≈1|/min). The transducer is not susceptible to interference from anesthetic agents, water vapor, nitrous oxide, carbon dioxide, or other gases and vapors apt to be present in the environment in which the system is used.
The sensor comprises a polymeric membrane in which a luminescable composition such as a porphyin dye is dispersed. The sensor membrane is the mediator that brings about dye-oxygen interaction in a controlled fashion. In a functional sensor, the dye is dispersed in the polymeric membrane, and oxygen diffuses through the polymer. The characteristics of the sensor are dependent upon the dye-polymer interaction and permeability and the solubility of oxygen in the polymer. Such characteristics include the sensitivity of response of the sensor to oxygen, the response time of the sensor to a change in oxygen concentration, and the measured values of phosphorescence intensity and decay time. Thus the composition and molecular weight of the polymer determines the sensor characteristics. Also, if the sensor is prepared by evaporation of a solution as described in the copending applications, the film characteristics depend on the solvent that is used and conditions during casting or evaporation. If the dye is separately doped into the film from another solution, the solvent and conditions in the doping medium also affect the sensor characteristics. When the polymer film is prepared by polymerization of a monomer or mixture, the sensor characteristics depend on the conditions of polymerization and such resultant polymer characteristics as degree of crosslinking and molecular weight.
The luminescent chemical sensor is not toxic to the patient and is a part of a consumable (i.e., disposable) airway adapter weighing less than 0.5 ounce. The sensor shelf life is greater than one year and the operational life exceeds 100 hours. The cost of the consumable airway adapter is minimal.
It is also important that the oxygen monitoring systems disclosed in the copending applications have sufficient accuracy (1.0%), precision (0.01%), and response time (<100 ms) to monitor breath-by-breath oxygen concentrations. The sensor is not sensitive to other gases found in the airway, including anesthetic agents, and is accordingly not excited into luminescence by those gases. The sensitivity of the sensor to temperature, flow rate, pressure and humidity change is well understood, and algorithms which provide compensation for any errors due to these changes are incorporated in the signal processing circuits of the device.
The visible light oxygen measurement transducers disclosed in the copending applications employ a sensor heater arrangement and a proportional-integrated-differential (PID) heater control system for keeping the oxygen concentration sensor of the transducer precisely at a selected operating temperature. This is particularly significant because those oxygen measurement transducers employ a sensor which involves the use of the diffusion of oxygen into a luminescable layer in measuring oxygen concentration. The rate of diffusion is temperature dependent. As a consequence, the measurement of oxygen concentration becomes inaccurate unless the sensor temperature is kept constant. Also, if the window through which the excitation energy passes is not kept warm, it may fog over. This also affects the accuracy of the oxygen concentration measurement.
The location of the oxygen concentration sensor in a replaceable, simple component is a feature of the systems disclosed in the copending applications. This makes it possible to readily and inexpensively ensure that the system is sterile with respect to each patient being monitored by replacing the airway adapter between patients, avoiding the non-desirability (and perhaps the inability) to sterilize that system component.
The provision of an airway adapter sensor and a separate signal-producing transducer also has the practical advantage that a measurement of oxygen concentration can be made without interrupting either the ventilation of a patient or any other procedure involving the use of the airway circuit. This is effected by installing the airway adapter in the airway circuit. When the time comes to make oxygen measurements, all that is required is that the transducer be coupled to the airway adapter already in place.
Another important feature of the invention ensures that the airway adapter and transducer are assembled in the correct orientation and that the airway adapter and transducer are securely assembled until deliberately separated by the system user.
The signals generated by the oxygen-measurement transducers of the previously disclosed system are processed to remove noise and extract the luminescence decay time, which is the oxygen-sensitive parameter of interest. A lock-in amplifier is preferably employed for this purpose. The lock-in amplifier outputs a signal which has a phase angle corresponding to the decay time of the excited, luminescent composition in the oxygen concentration sensor. The lock-in detection circuitry rejects noise and those components of the photodetector-generated signal which are not indicative of oxygen concentration. This noise reduction also allows a higher level of signal gain which, in turn, makes possible enhanced measurement precision while decreasing the level of the visible excitation. This reduces instability from photoaging of the sensor, increasing accuracy and useable life. All of this processing, which can be done with digital, analog, or hybrid method, is fast enough for even the most demanding applications such as those requiring the breath-by-breath monitoring of a human patient. Various pathological conditions result in a change of oxygen demand by the body. If a decrease of oxygen utilization by the body, for example, can be detected on a breath-by-breath basis, timely and effective remedial steps can be taken to assist the patient.
In the novel oxygen measurement transducers of the present invention, the concentration of oxygen in the gases being monitored is reflected in the quenching of an excited luminescent composition in the oxygen concentration sensor by oxygen diffusing into the sensor matrix. A source consisting of a light-emitting diode (LED) produces visible exciting light which strikes the surface of the sensor film. Some of the light is absorbed by the luminescent chemical dye in the film whereupon it produces luminescent light at a second, shifted wavelength. This light is captured by a photodetector which thereupon generates a signal reflecting the intensity and decay pattern of the intercepted light. All light directed toward the photodetector can potentially result in a signal. A suitable optical filter placed over the surface of the photodetector discriminates against all but the luminescent light, thereby ensuring that the photodetector is producing a signal related to oxygen concentration only.